Skip to content

Upgrade sqlparser-rs to 0.51.0, support new interval logic from sqlparse-rs #12222

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Sep 15, 2024
Merged

Upgrade sqlparser-rs to 0.51.0, support new interval logic from sqlparse-rs #12222

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
179be76
support new interval logic from sqlparse-rs
samuelcolvin Aug 28, 2024
edf318b
uprev sqlparser-rs branch
samuelcolvin Aug 29, 2024
ab88318
Merge branch 'main' into sqlparse-interval
samuelcolvin Sep 12, 2024
6bba073
use sqlparser 51
samuelcolvin Sep 12, 2024
268d478
better extract logic and interval testing
samuelcolvin Sep 12, 2024
8fe6a2f
revert unnecessary changes
samuelcolvin Sep 12, 2024
5ad18e0
revert unnecessary changes, more
samuelcolvin Sep 12, 2024
e7ed21e
cleanup
samuelcolvin Sep 12, 2024
ce3425a
fix last failing test :fingerscrossed:
samuelcolvin Sep 12, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,7 +134,8 @@ rand = "0.8"
regex = "1.8"
rstest = "0.22.0"
serde_json = "1"
sqlparser = { version = "0.50.0", features = ["visitor"] }
#sqlparser = { version = "0.50.0", features = ["visitor"] }
sqlparser = { git = "https://github.com/samuelcolvin/sqlparser-rs", rev = "31a19f7", features = ["visitor"] }
tempfile = "3"
thiserror = "1.0.44"
tokio = { version = "1.36", features = ["macros", "rt", "sync"] }
Expand Down
6 changes: 2 additions & 4 deletions datafusion-cli/Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

4 changes: 1 addition & 3 deletions datafusion/sql/src/expr/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -197,9 +197,7 @@ impl<'a, S: ContextProvider> SqlToRel<'a, S> {
}

SQLExpr::Array(arr) => self.sql_array_literal(arr.elem, schema),
SQLExpr::Interval(interval) => {
self.sql_interval_to_expr(false, interval, schema, planner_context)
}
SQLExpr::Interval(interval) => self.sql_interval_to_expr(false, interval),
SQLExpr::Identifier(id) => {
self.sql_identifier_to_expr(id, schema, planner_context)
}
Expand Down
2 changes: 1 addition & 1 deletion datafusion/sql/src/expr/unary_op.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,7 +43,7 @@ impl<'a, S: ContextProvider> SqlToRel<'a, S> {
self.parse_sql_number(&n, true)
}
SQLExpr::Interval(interval) => {
self.sql_interval_to_expr(true, interval, schema, planner_context)
self.sql_interval_to_expr(true, interval)
}
// not a literal, apply negative operator on expression
_ => Ok(Expr::Negative(Box::new(self.sql_expr_to_logical_expr(
Expand Down
195 changes: 72 additions & 123 deletions datafusion/sql/src/expr/value.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ use datafusion_expr::expr::{BinaryExpr, Placeholder};
use datafusion_expr::planner::PlannerResult;
use datafusion_expr::{lit, Expr, Operator};
use log::debug;
use sqlparser::ast::{BinaryOperator, Expr as SQLExpr, Interval, Value};
use sqlparser::ast::{BinaryOperator, Expr as SQLExpr, Interval, UnaryOperator, Value};
use sqlparser::parser::ParserError::ParserError;
use std::borrow::Cow;

Expand Down Expand Up @@ -168,12 +168,11 @@ impl<'a, S: ContextProvider> SqlToRel<'a, S> {

/// Convert a SQL interval expression to a DataFusion logical plan
/// expression
#[allow(clippy::only_used_in_recursion)]
pub(super) fn sql_interval_to_expr(
&self,
negative: bool,
interval: Interval,
schema: &DFSchema,
planner_context: &mut PlannerContext,
) -> Result<Expr> {
if interval.leading_precision.is_some() {
return not_impl_err!(
Expand All @@ -196,127 +195,42 @@ impl<'a, S: ContextProvider> SqlToRel<'a, S> {
);
}

// Only handle string exprs for now
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is great to remove all this stuff from the sql planner (as it is now in the parser)

let value = match *interval.value {
SQLExpr::Value(
Value::SingleQuotedString(s) | Value::DoubleQuotedString(s),
) => {
if negative {
format!("-{s}")
} else {
s
}
}
// Support expressions like `interval '1 month' + date/timestamp`.
// Such expressions are parsed like this by sqlparser-rs
//
// Interval
// BinaryOp
// Value(StringLiteral)
// Cast
// Value(StringLiteral)
//
// This code rewrites them to the following:
//
// BinaryOp
// Interval
// Value(StringLiteral)
// Cast
// Value(StringLiteral)
SQLExpr::BinaryOp { left, op, right } => {
let df_op = match op {
BinaryOperator::Plus => Operator::Plus,
BinaryOperator::Minus => Operator::Minus,
BinaryOperator::Eq => Operator::Eq,
BinaryOperator::NotEq => Operator::NotEq,
BinaryOperator::Gt => Operator::Gt,
BinaryOperator::GtEq => Operator::GtEq,
BinaryOperator::Lt => Operator::Lt,
BinaryOperator::LtEq => Operator::LtEq,
_ => {
return not_impl_err!("Unsupported interval operator: {op:?}");
}
};
match (
interval.leading_field.as_ref(),
left.as_ref(),
right.as_ref(),
) {
(_, _, SQLExpr::Value(_)) => {
let left_expr = self.sql_interval_to_expr(
negative,
Interval {
value: left,
leading_field: interval.leading_field.clone(),
leading_precision: None,
last_field: None,
fractional_seconds_precision: None,
},
schema,
planner_context,
)?;
let right_expr = self.sql_interval_to_expr(
false,
Interval {
value: right,
leading_field: interval.leading_field,
leading_precision: None,
last_field: None,
fractional_seconds_precision: None,
},
schema,
planner_context,
)?;
return Ok(Expr::BinaryExpr(BinaryExpr::new(
Box::new(left_expr),
df_op,
Box::new(right_expr),
)));
}
// In this case, the left node is part of the interval
// expr and the right node is an independent expr.
//
// Leading field is not supported when the right operand
// is not a value.
(None, _, _) => {
let left_expr = self.sql_interval_to_expr(
negative,
Interval {
value: left,
leading_field: None,
leading_precision: None,
last_field: None,
fractional_seconds_precision: None,
},
schema,
planner_context,
)?;
let right_expr = self.sql_expr_to_logical_expr(
*right,
schema,
planner_context,
)?;
return Ok(Expr::BinaryExpr(BinaryExpr::new(
Box::new(left_expr),
df_op,
Box::new(right_expr),
)));
}
_ => {
let value = SQLExpr::BinaryOp { left, op, right };
return not_impl_err!(
"Unsupported interval argument. Expected string literal, got: {value:?}"
);
}
if let SQLExpr::BinaryOp { left, op, right } = *interval.value {
let df_op = match op {
BinaryOperator::Plus => Operator::Plus,
BinaryOperator::Minus => Operator::Minus,
_ => {
return not_impl_err!("Unsupported interval operator: {op:?}");
}
}
_ => {
return not_impl_err!(
"Unsupported interval argument. Expected string literal, got: {:?}",
interval.value
);
}
};
};
let left_expr = self.sql_interval_to_expr(
negative,
Interval {
value: left,
leading_field: interval.leading_field.clone(),
leading_precision: None,
last_field: None,
fractional_seconds_precision: None,
},
)?;
let right_expr = self.sql_interval_to_expr(
false,
Interval {
value: right,
leading_field: interval.leading_field,
leading_precision: None,
last_field: None,
fractional_seconds_precision: None,
},
)?;
return Ok(Expr::BinaryExpr(BinaryExpr::new(
Box::new(left_expr),
df_op,
Box::new(right_expr),
)));
}

let value = interval_literal(*interval.value, negative)?;

let value = if has_units(&value) {
// If the interval already contains a unit
Expand All @@ -343,6 +257,41 @@ impl<'a, S: ContextProvider> SqlToRel<'a, S> {
}
}

fn interval_literal(interval_value: SQLExpr, negative: bool) -> Result<String> {
let s = match interval_value {
SQLExpr::Value(Value::SingleQuotedString(s) | Value::DoubleQuotedString(s)) => s,
SQLExpr::Value(Value::Number(ref v, long)) => {
if long {
return not_impl_err!(
"Unsupported interval argument. Long number not supported: {interval_value:?}"
);
} else {
v.to_string()
}
}
SQLExpr::UnaryOp { op, expr } => {
let negative = match op {
UnaryOperator::Minus => !negative,
UnaryOperator::Plus => negative,
_ => {
return not_impl_err!(
"Unsupported SQL unary operator in interval {op:?}"
);
}
};
interval_literal(*expr, negative)?
}
_ => {
return not_impl_err!("Unsupported interval argument. Expected string literal or number, got: {interval_value:?}");
}
};
if negative {
Ok(format!("-{s}"))
} else {
Ok(s)
}
}

// TODO make interval parsing better in arrow-rs / expose `IntervalType`
fn has_units(val: &str) -> bool {
let val = val.to_lowercase();
Expand Down
44 changes: 35 additions & 9 deletions datafusion/sql/tests/cases/plan_to_sql.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ use datafusion_sql::unparser::dialect::{
use datafusion_sql::unparser::{expr_to_sql, plan_to_sql, Unparser};

use datafusion_functions::core::planner::CoreFunctionPlanner;
use sqlparser::dialect::{Dialect, GenericDialect, MySqlDialect};
use sqlparser::dialect::{Dialect, GenericDialect, MySqlDialect, PostgreSqlDialect};
use sqlparser::parser::Parser;

use crate::common::{MockContextProvider, MockSessionState};
Expand Down Expand Up @@ -481,11 +481,17 @@ fn test_table_references_in_plan_to_sql() {
assert_eq!(format!("{}", sql), expected_sql)
}

test("catalog.schema.table", "SELECT catalog.\"schema\".\"table\".id, catalog.\"schema\".\"table\".\"value\" FROM catalog.\"schema\".\"table\"");
test("schema.table", "SELECT \"schema\".\"table\".id, \"schema\".\"table\".\"value\" FROM \"schema\".\"table\"");
test(
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is a driveby (unrelated) cleanup, right? (this is fine I am just verifying my understanding)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It was necessary to get tests to pass, I assume it came from some other change in sqlparser, I didn't look too hard.

But it's not just cosmetic.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this mostly changes catalog to "catalog"
a driveby is usage or raw literals (which is a nice thing on its own)

"catalog.schema.table",
r#"SELECT "catalog"."schema"."table".id, "catalog"."schema"."table"."value" FROM "catalog"."schema"."table""#,
);
test(
"schema.table",
r#"SELECT "schema"."table".id, "schema"."table"."value" FROM "schema"."table""#,
);
test(
"table",
"SELECT \"table\".id, \"table\".\"value\" FROM \"table\"",
r#"SELECT "table".id, "table"."value" FROM "table""#,
);
}

Expand All @@ -507,10 +513,10 @@ fn test_table_scan_with_no_projection_in_plan_to_sql() {

test(
"catalog.schema.table",
"SELECT * FROM catalog.\"schema\".\"table\"",
r#"SELECT * FROM "catalog"."schema"."table""#,
);
test("schema.table", "SELECT * FROM \"schema\".\"table\"");
test("table", "SELECT * FROM \"table\"");
test("schema.table", r#"SELECT * FROM "schema"."table""#);
test("table", r#"SELECT * FROM "table""#);
}

#[test]
Expand Down Expand Up @@ -590,8 +596,8 @@ fn test_pretty_roundtrip() -> Result<()> {
Ok(())
}

fn sql_round_trip(query: &str, expect: &str) {
let statement = Parser::new(&GenericDialect {})
fn sql_round_trip(query: &str, expect: &str, dialect: &dyn Dialect) {
let statement = Parser::new(dialect)
.try_with_sql(query)
.unwrap()
.parse_statement()
Expand All @@ -609,16 +615,36 @@ fn sql_round_trip(query: &str, expect: &str) {

#[test]
fn test_interval_lhs_eq() {
sql_round_trip(
"select interval 2 second = interval 2 second",
"SELECT (INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS' = INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS')",
&GenericDialect {}
);
}

#[test]
fn test_interval_lhs_eq_pg() {
sql_round_trip(
"select interval '2 seconds' = interval '2 seconds'",
"SELECT (INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS' = INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS')",
&PostgreSqlDialect {}
);
}

#[test]
fn test_interval_lhs_lt() {
sql_round_trip(
"select interval 2 second < interval 2 second",
"SELECT (INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS' < INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS')",
&GenericDialect {}
);
}

#[test]
fn test_interval_lhs_lt_pg() {
sql_round_trip(
"select interval '2 seconds' < interval '2 seconds'",
"SELECT (INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS' < INTERVAL '0 YEARS 0 MONS 0 DAYS 0 HOURS 0 MINS 2.000000000 SECS')",
&PostgreSqlDialect {}
);
}
Loading